Stories
Slash Boxes
Comments

News for nerds, stuff that matters

Slashdot Log In

Log In

Create Account  |  Retrieve Password

The Men Who Fix the Internet

Posted by kdawson on Mon Mar 16, 2009 10:36 PM
from the physical-side-of-things dept.
An anonymous reader writes "Remember all those undersea cables breaking? PopSci.com introduces John Rennie, who '... has braved the towering waves of the North Atlantic Ocean to keep your e-mail coming to you. As chief submersible engineer aboard the Wave Sentinel, part of the fleet operated by UK-based undersea installation and maintenance firm Global Marine Systems, Rennie — a congenial, 6'4", 57-year-old Scotsman — patrols the seas, dispatching a remotely operated submarine deep below the surface to repair undersea cables.' The article goes on to outline the physical infrastructure of the Internet, including some of its points of vulnerability."
+ -
story

Related Stories

[+] Fourth Undersea Cable Taken Offline In Less Than a Week 499 comments
An anonymous reader writes "Another undersea cable was taken offline on Friday, this one connecting Qatar and UAE. 'The [outage] caused major problems for internet users in Qatar over the weekend, but Qtel's loss of capacity has been kept below 40% thanks to what the telecom said was a large number of alternative routes for transmission. It is not yet clear how badly telecom and internet services have been affected in the UAE.' In related news it's been confirmed that the two cables near Egypt were not cut by ship anchors." Update: 02/04 07:13 GMT by Z : A commenter notes that despite the language in the article indicated a break or malfunction, the cable wasn't cut. It was taken offline due to power issues.
[+] IT: Fifth Cable Cut To Middle East 676 comments
You may have noticed a number of stories recently about undersea cables getting cut around the world. Apparently the total is now up to 5, but the scariest part of this is that Iran is now offline. You can also read Schneier's comments on this coincidence. Update: 02/06 17:42 GMT by Z : As a commenter notes, though the country of Iran is obviously experiencing some networking difficulties, it is not offline.
[+] Mediterranean Undersea Cables Cut, Again 329 comments
miller60 writes "Three undersea cables in the Mediterranean Sea have failed within minutes of each other in an incident that is eerily similar to a series of cable cuts in the region in early 2008. The cable cuts are already causing serious service problems in the Middle East and Asia. See coverage at the Internet Storm Center, Data Center Knowledge and Bloomberg. The February 2008 cable cuts triggered rampant speculation about sabotage, but were later attributed to ships that dropped anchor in the wrong place."
This discussion has been archived. No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
 Full
 Abbreviated
 Hidden
More
Loading... please wait.
  • I suggest we leave them that way. It will reduce spam, and make Dell hire locally for their call centers.

  • by Starteck81 (917280) on Monday March 16 2009, @10:44PM (#27221199)
    Grounds keeper Willie of the undersea cables, at your service.
  • by ColdWetDog (752185) * on Monday March 16 2009, @10:47PM (#27221217) Homepage

    The cables regularly fail. On any given day, somewhere in the world there is the nautical equivalent of a hit and run when a cable is torn by fishing nets or sliced by dragging anchors. If the mishap occurs in the Irish Sea, the North Sea or the North Atlantic, Rennie comes in to splice the break together.

    WTF are people dragging anchors around for? I would presume (and could be entirely wrong, as usual) that shallow water cable runs wouldn't be located next to anchorages. Do these sea going vessels have to stop for lunch or something?

    And why to we even allow fisherman to drag crap along the sea bottom? I thought industrial level trawling went out years ago?

  • Because anyone can still whoop Aquaman's butt.

    • Re: (Score:2, Funny)

      by Anonymous Coward

      Because anyone can still whoop Aquaman's butt.

      No, it's because they keep the flow of pr0n running.

  • by Anonymous Coward on Monday March 16 2009, @10:59PM (#27221289)
    While there is loads of critical data that goes through these cables, I feel bad when these guys are working their asses off to make sure that 4chan or youtubes of a chimpanzee riding on a segway gets to its proper place.
  • by dwhitaker (1500855) on Monday March 16 2009, @11:15PM (#27221373) Homepage
    This is yet another example of the jobs which we rely on everyday but don't give much thought to. Also, this make me really think there is a great job out there to fit everyone. (When the economy improves that is.)
  • by martin-boundary (547041) on Monday March 16 2009, @11:21PM (#27221419)
    Gimme Terabit factor nine! Kirk to Engineering, I need more downloads, Scotty!

    Aye Captain, but I don't know if my poor cables will take more.

  • by Clancie (678344) on Monday March 16 2009, @11:44PM (#27221555)
    Holy crap! The Internet *is* a series of tubes! Evidence:Image from TFA [popsci.com]
  • by trawg (308495) on Tuesday March 17 2009, @12:37AM (#27221815) Homepage

    ...is a cool article up on Wired [wired.com] (look for the printable link option so it's all on one page) detailing an interesting adventure around the world and some of the history of undersea cables. Definitely worth a read.

    • Re: (Score:3, Informative)

      This article [wikipedia.org] details the building of the first transatlantic cables in the 1850s & 1860s. Definitely trickier to repair back then (lay another one)...
  • Vulnerability? (Score:3, Interesting)

    by AHuxley (892839) on Tuesday March 17 2009, @12:38AM (#27221821)
    "The article goes on to outline the physical infrastructure of the Internet, including some of its points of vulnerability"

    Sean Gorman mapped out the US fiber-optic telco fiefdoms.
    Parts of his dissertation where "removed".
    http://www.wired.com/science/discoveries/news/2006/01/70040?currentPage=2 [wired.com]
    Getting back to the popsci 'news'
    The part I find interesting is the use of 'hubs'
    Are hubs (fiber locations?) for cost savings, lazy design, best design for a shareholder when burning tax payers re nation building, collusion between telcos, easy NSA access ?
    What do other parts of the world do ?
    • Re:Vulnerability? (Score:4, Informative)

      by tqft (619476) <ianburrows_au@NOsPAM.yahoo.com> on Tuesday March 17 2009, @02:58AM (#27222365) Homepage Journal

      "What do other parts of the world do ?"
      The same thing

      "Are hubs (fiber locations?) for cost savings, lazy design, best design for a shareholder when burning tax payers re nation building, collusion between telcos, easy NSA access ?"
      All of the above
      At some point you need to connect network E from Elbonia to network P from PHiliBelphia and also networks a through z. This starts to get expensive real fast no matter how you do it. Doing it in one place lowers cost (hub) but focuses for a point of failure. As the article said - you can't get away from this. I like how they said the best way to prevent cascading network damage is to shutdown the "nearest" hub connections to the failed point to minimize the damage - like they do when the electricity transmission network or a generator goes off somewhere. It isn't optional and can't really be worked around - if you want random person E to get stuff from P then you need interconnects somewhere. Yeah you could do it with a lot (a real lot) of little interconnects all over the place - just don't and service them as your staff will always be in the field.

  • by ItaliaMatt (581886) on Tuesday March 17 2009, @02:08AM (#27222155)

    Everytime we have a connectivity hiccup I am flooded with calls from our users asking "Is the Internet broken?"

    It takes everything in my power not to say "Yes. The Internet is, in fact, broken"

  • Your doing it wrong (Score:5, Interesting)

    by FridgeFreezer (1352537) on Tuesday March 17 2009, @05:11AM (#27222825)

    From TFA:

    "If terrorists managed to gain remote access to a facility's command-and-control system, they could, for example, cause the generators to overheat and explode."

    If you can make a generator explode on command, you really are doing it wrong. Backup generators may be able to be remotely started, stopped, switched in/out and checked but you should not be able to do the equivalent of burnouts with them.

    Additionally, the article states that catastrophic failures would start to creep in after ~2 days of no human maintenance. WTF? Most exchanges and data centres I've been in are ghost ships 350 days a year aside from upgrades and config changes, how is it that such critical hardware can't tick over by itself for a month or so without going nipples skyward?

    Hell, the average telephone exchange, if you nuked everything around it, would be giving dialtone and DSL to the skeletons for at least a week, probably more depending on how much diesel is in the tanks.

    • by steelfood (895457) on Tuesday March 17 2009, @09:46AM (#27225163)

      Hell, the average telephone exchange, if you nuked everything around it, would be giving dialtone and DSL to the skeletons for at least a week, probably more depending on how much diesel is in the tanks.

      During the northeast blackout of '03, all of the utilities went out. Street lights, traffic signals, all went dark. Cell phone towers went out. There wasn't enough water pressure beyond 5 stories. There was gas, but no starter. But there was still a dialtone through the landline. And we could still make calls out, if our phone didn't require an external power source. Most of the handsets we use are cordless, but we have several of the simpler phones lying around for such emergencies.

    • by Shikaku (1129753) on Monday March 16 2009, @11:13PM (#27221355)

      Latency is a huge problem with that idea buddy.

    • by fractoid (1076465) on Monday March 16 2009, @11:14PM (#27221361) Homepage
      There are enough satellites up there that we can get *some* communications without cables. Those satellite links suck at the best of times, though - if nothing else they have horrible latency, and can't approach the huge bandwidth of an undersea cable full of optic fibres. Just like in your own apartment, wireless is cool for convenience and for when you have a kitten (or fishing trawler) messing with your wiring, but cabling is always faster and better for fixed installations.
    • by CharlieG (34950) on Monday March 16 2009, @11:28PM (#27221467) Homepage

      One HUGE propblem with satcoms, and why satcom pretty much went away for telephone - latency. A geosync sat orbits at 26200 miles (roughly), making a 2 way trip (up and down) a 52400 mile trip from point A to point B, or making it take a tad over .28 seconds. Now wait for your ACK to come back, another .28 seconds, and think about what you have. A slow, limited bandwith link. Generally, Satcoms have become used in one way "broadcast" type trasmissions (send the 30 minute TV show up, don't worry about the 1/4 second, as they are recording on the toher side) OR "Ad-hoc" communications, where you don't KNOW where the other station will be (a ship on the ocean, a TV news crew that is in Akron today, and Iowa next week, and even then, they try to keep the signals off the birds unless it's breaking news. You try to get it to a local affiliate, and land line it back)

      • by Brickwall (985910) on Tuesday March 17 2009, @12:40AM (#27221829)
        Iridium puts its satellites in low earth orbits to avoid the latency issue. It's geosynch orbits that are up high. But I do remember making telephone calls that went over geosynch satellite back in the 80's, and the latency is really annoying - you're never quite sure when the other person is finished talking, so you end up talking over each other, and having awkward pauses. When fibre became common, most telcos stopped using satellite.
        • by CharlieG (34950) on Tuesday March 17 2009, @01:01AM (#27221925) Homepage

          Correct. I forgot to bring up the Low Earth Orbit Sat Phones (aka Iridium), which is another kettle of fish. The big problem there is limited channels, again, your not going to have the kind of bandwith you need for serious internet (note, I said serious, like multiple OC3 stuff).

          Interestingly, NATO, with all their Sats, and Iridium (Remember, the US Military basically keeps them in business) is re-looking at HF radio comms. Ultra high speed 24 bit DSPs, and other technologies are making them clearer and more reliable (less dependent on operator skill), and they have the advantage of working when you have a limited sky view

          • Why don't they get with the times and Internet it?

            Oh, they do even better - they now have holograms.
    • by Anonymous Coward on Monday March 16 2009, @11:32PM (#27221483)

      Oh those kids! Never had to work with a megabit satellite link connected somewhere in Africa and try to send VoIP to America, haven't you?
      The latency is not the only problem, there are magnetic storms, other satellites crossing into the sight of yours, bad weather, and so much other crap that I can't even remember.
      That is why we need thick undersea cables or all your beautiful iPhones and other gadgets will be rendered totally useless...

    • by fuzzyfuzzyfungus (1223518) on Monday March 16 2009, @11:17PM (#27221393) Journal
      I might be walking into a woosh here; but radio waves are (or rather carry) data that travels at the speed of light(plus, the speed of light is higher in a vacuum than in fiber). What really kills you with satellites is the distance.
      • Re: (Score:3, Interesting)

        Satellites are cost effective if you are either:
        1. Reaching a broad audience with the same transmission.
        2. A large government with cryptic and voluminous bookkeeping designed to hide that you are at a loss.

        Just ask the satellite phone companies what happens when you have to listen to that broad audience too.

      • by Muad'Dave (255648) on Tuesday March 17 2009, @08:11AM (#27223953) Homepage
        You are correct, grasshopper. The speed at which light or radio waves propagate in a medium is determined by that medium's index of refraction, which (as you stated) is 1 for a vacuum. A typical value for fiber is 1.47, so the signal in the fiber travels only 68% as fast as the radio wave in a vacuum. The distances involved with geosynchronous orbits, however, make up for that slower speed. In the time it takes the radio wave to reach the satellite (35,786 km / 300,000 km/s = 119 mS), the fiber signal could've circled the Earth 1.2 times (119 mS * 204,082 km/s = 24,286 km (Diameter of the Earth is approx 20,000 km).
    • You beat me to it, but I agree that Stephenson's piece is fascinating reading. Someone with points please mod parent up.
    • That article is a classic well worth reading.

      I actually subscribed to Wired at the time based on that article. Sadly, their regular content was nowhere near as good.